The Agent Skills Directory

[PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection as it ingests untrusted data from service logs and allows for the execution of arbitrary SQL queries.
Ingestion points: Log content retrieved via the list_logs tool and SQL strings passed to query_render_postgres (found in SKILL.md).
Boundary markers: The instructions do not define specific delimiters or warnings to ignore instructions embedded within the log data.
Capability inventory: The skill has the ability to execute database queries (query_render_postgres), read external resource logs (list_logs), and access service environments via the CLI (render ssh).
Sanitization: There is no evidence of input validation or sanitization for the SQL queries or log data processed by the agent.
[COMMAND_EXECUTION]: The skill utilizes the Render CLI for several administrative operations, including render ssh for remote terminal access and render psql for database connectivity. These tools provide significant capabilities within the target service environment as part of the intended monitoring and debugging workflow.
[EXTERNAL_DOWNLOADS]: The skill references the official Render MCP server endpoint (https://mcp.render.com/mcp) for setup. This is a reference to a well-known service provider's infrastructure and is considered a safe operational requirement.

render-monitor